Analysis of the factors affecting mortality in patients followed in our pediatric intensive care unit

Introduction:Pediatric intensive care units (PICU) are specialized units where children with life threatening conditions are followed and treated. These units aim to provide treatment for and close follow-up of patients with life threatening diseases, which are possible only in intensive care environment. As the clinical conditions of patients admitted to PICU may differ greatly from each other, it is deemed essential to compare patients in a given PICU and the PICU department as regards morbidity and mortality. Determining the factors contributingto mortality may prove beneficial for the determination of possible prognoses and theapplication of treatment methods. Materials and Methods:This study was carried out in the PICU of Abant Izzet Baysal University Faculty of Medicine Training and Research Hospital. The medical records of patients treated in our unit between thedates of 1 January 2014 and 31 December 2015 were retrospectively analyzed as regards mortality and prognostic. Results:The number of girls and boys did not significantlydiffer from each other, 46.9% and 53.1%, respectively (p>0.05). The mean age of patients discharged from our PICU was 61 months, whereas the mean age of dead patients was 78 months, which indicates there was no significant differencebetween the two groups as regards mean age (p=0.178). The duration of stay inthe PICU ranged from 1 day to 42 days and the average duration of stay was calculated as 11±4.41 days. The patients discharged and patients dead during their treatment at the PICU (p=0.005) significantly differed as regards average duration of stay (4.00±4.33 days and 6.88±5.51 days, respectively). Regression analysis showed that the duration of stay in the PICU was an important factor for mortality, and the longer a patient stayed in the PICU, the higher the risk of mortality (p=0.004). The most common reason for admittance in the PICU was accidents-poisoning (33.7%), followed by respiratory system (28.6%), central nervous system (22.6%), cardiovascular system (4.7%), gastrointestinal system (1.6%), and hematologic-oncologic diseases (1%). The vii comparison of blood gas parameters of discharged and dead patients significantly differed in pH (7.35±0.09 and 7.24±0.18, respectively, p= 0.004), HCO3(20.76±4.83 and 16.41±7.37, respectively, p= 0.004), and lactate (2.11±1.82 and 4.41±4.22, respectively, p= 0.000) levels. Hyperlactatemia in particular was found to have a decisive effect on mortality (138.6 times, p=0.000). The comparison of discharged and dead patients as regards the giving of erythrocyte suspension by central venous catheterization, fresh frozen plasma, intravenous immunoglobuline, and platelet suspension (hematologic-oncologic patients) revealed significant differences (p=0.000). It was found that 48.5% of patients weretransferred to the relevant departments, 44.8% were discharged, 3.7% died, and 3% were transferred to other hospitals. Mortality in patients occurred due to the following causes: 52% cardiovascular, 44% pulmonary, and 4% neurological.Comparison of secondary diagnoses revealed significant differences between discharged and dead patients as regards sepsis (p=0.000), shock (p= 0.000), multiple organ failure (p=0.000), and renal failure (p=0.000). Neutrophil and platelet lymphocyte ratios were found to be higher in mortality cases. Neutrophil- lymphocyte ratio(NLR), platelet-lymphocyte ratio(PLR), and C-reactive protein (CRP)/mean platelet volume(MPV) values were higher in dead patients (NLRdischarged: 3.59±6.87, NLRdead: 5.87±5.72, p= 0.037; PLRdischarged: 10.58±34.99, PLRdead: 63.08±133.19, p=0.000; and CRP/MPVdischarged: 2.12±4.99, PLRdead: 7.08±12.03, p= 0.001). Conclusions:According to the clinical and laboratory findings as well as the results of the scoring methodology used (Pediatric LogisticOrgan Dysfunction, Glaskow Coma Scale, Pediatric Index of Mortality and Pediatric Risk of Mortality), mortality rates were found to be higher in patients with prolonged stay in the PICU, shock, sepsis, multiple organ failure, hemato-oncologic diseases, a need for central venous catheterisation, surgical operation, or mechanical ventilation, and patients who needed blood or blood components transfusion (p=0.000). These findings suggest that the use of scoring methodology in patients followed in PICU, along with clinical and laboratory findings may be useful in evaluatingmortality and organ failure.